Central nervous system (CNS) infections are common in Sub-Saharan Africa, either with or without HIV-infection across all ages. In persons with HIV, cryptococcal meningitis (CM) is the second most common AIDS defining illness in Africa, and now with the availability of HIV antiretroviral therapy (ART), long term survival should be possible. However, the new challenge of HIV immune reconstitution inflammatory syndrome (IRIS) has emerged. IRIS is a poorly understood immunologic phenomenon whereby a portion of persons (~30%) with AIDS starting ART paradoxically worsen as their immune systems improve. IRIS events are characterized by exaggerated inflammation in the setting of microbiologic treatment success. When the dysregulated inflammation of IRIS occurs in the CNS, death frequently occurs, yet the neurologic outcome among survivors is unknown. We propose a prospective cohort study of persons presenting with CNS infections in Sub- Saharan Africa with HIV-infection. We will use molecular diagnostics to determine the etiologies of CNS infections. After persons initiate ART, we will prospectively conduct surveillance for IRIS and assess neurological, functional, and neuro-cognitive status. We will profile cytokines in the cerebrospinal fluid (CSF) to discover biomarkers predictive of poor neurologic outcome, future IRIS, and/or death. Hypothesis: We hypothesize that persons with advanced HIV (CD4 <100) have worse neurologic outcomes, and persons with subsequent CNS-related IRIS events have worse neurologic outcomes than those who do not experience CNS-IRIS. We hypothesize that pro-inflammatory baseline cytokine profiles of the CNS will be predictive of future adverse outcomes. Specific Aims 1) We will determine the etiology and neurologic outcomes of CNS infections in Sub-Saharan Africa among adolescents, adults, and elderly with HIV-infection. 2) For AIDS patients who have IRIS-related CNS infections, we will determine their neurologic outcomes after they initiate antiretroviral therapy (ART) in order to determine if patients who develop Immune Reconstitution Inflammatory Syndrome (IRIS) have worse outcomes compared to those who do not develop IRIS. 3) We will determine whether specific CSF cytokine profiles can predict worse neurological outcomes in patients with CNS infections or predict IRIS in patients with CNS infections and AIDS. PUBLIC HEALTH RELEVANCE: Cryptococcal meningitis (CM) is the second most common AIDS defining illness in Sub- Saharan Africa causing 30% of the AIDS-attributable mortality in Africa. Other CNS infections also occur, including aseptic meningitis of unknown etiology. With the availability of antiretroviral therapy (ART), the new challenge of Immune Reconstitution Inflammatory Syndrome (IRIS) occurs resulting in paradoxical clinical deterioration and mortality. The impact of IRIS on neurologic outcomes is unknown. Biomarkers to predict IRIS are needed.